X-DBER 2021 Conference

To establish a viable community of practice, X-DBER researchers need to identify and collaborate on avenues of research in which a cross-disciplinary approach can yield insights that would not arise from a single discipline. We propose that feedback loops (FLs) are one such area of research. Feedback loops are systems in which an initial action triggers a chain of influences that either amplifies or counteracts the initial action. FLs have explanatory, predictive, and solution-shaping power across multiple domains, including climate, ecology, epidemiology, physiology, marketing, sociology, and engineering. Our research suggests that reasoning about FLs in multiple disciplinary contexts helps build the general FL concept through the process of mutual alignment analogy more effectively than learning about a single FL in a single discipline. FL systems that span disciplinary boundaries can support interdisciplinary collaboration, insofar as what person A cares about is both the cause and the consequence of what person B cares about. FLs pose interesting cognitive challenges, requiring the reasoner to accept that causality can loop backward, even though time cannot. A collaborative, cross-disciplinary research push to understand how learners and experts comprehend and reason with feedback loops could help build a populace able to (a) recognize FLs when they encounter them in unfamiliar contexts, (b) deploy FL understanding to explain and anticipate behaviors of growth, decay or stability, and (c) leverage FLs in designing solutions to some of humanity’s knottiest problems. Please reach out to us if this is a research direction you would like to pursue.

Dialogic Online Videos in STEM Learning 

Despite the tremendous growth in online mathematics and science videos, the dominant model is still a talking head or hand presenting a lecture or solving tasks. In contrast, alternatives have begun to emerge that feature dialogue, often between a tutor and an undergraduate, in biology, physics, and computer literacy. We were inspired by such work to create a model for dialogically-intensive, conceptually-oriented math videos for secondary school students (available at https://mathtalk.sdsu.edu/wordpress/). Each video features the unscripted dialogue of a pair of secondary school students as they convey sources of confusion and resolve their own dilemmas by arguing for and against particular ways of reasoning. Other students who learn by viewing and engaging with the videos are called vicarious learners (because they are participating in the original dialogue indirectly). This talk highlights research studies on vicarious learning – from our project and from other STEM disciplines. I also discuss theory regarding the foundations, benefits, and constraints of vicarious learning.

Part of the struggle learners face when confronting concepts in molecular genetics comes from the very nature of molecular cell biology (MCB) knowledge itself. Molecular mechanisms rely on multi-level reasoning, which is very difficult for beginners. Concepts and processes of MCB are “un-seeable” and rely on the ability to create and correctly interpret visual representations. Undergraduate students must develop skills of visual literacy, but to date there is no unified framework that describes common visual representations of MCB. We developed a 3x3 matrix, “the DNA Landscape,” based on examination of diagrams in undergraduate textbooks. The DNA Landscape is both a research and teaching tool that recognizes two dimensions of representations for any diagram involving DNA: scale (nucleotide through chromosomal levels) and abstraction (how closely it resembles the actual shape of the molecule). We tested the robustness of the Matrix by coding >2000 figures from 12 textbooks. Many figures contained multiple representations, but all were able to be coded using this new framework. Different MCB topics tended to use specific types of representations. For example, Mendelian genetics uses allele names, scale maps, and chromosomes quite often, but gene expression tends toward box and line structures and chromatin “string.” Preliminary results suggest that experts move easily around the landscape, but students have difficulty making connections between different representations of the same phenomena. To help learners develop an integrated knowledge base, educators need to go beyond assessing individual points within the matrix and instead teach and assess the connections between points.

Advancements in organic chemistry depend upon chemists’ ability to interpret NMR spectra, though research demonstrates that cultivating such proficiency requires years of graduate-level study. The organic chemistry community thus needs insight into how this expertise develops to expedite learning among its newest members. This study investigated undergraduate and doctoral chemistry students’ understanding and information processing during the interpretation of 1H NMR spectra and complementary IR spectra. Eye movements were measured to identify differences in cognitive processes between undergraduate and doctoral participants, and interviews were conducted to elucidate the assumptions that guided participants’ reasoning. Results suggest five areas of understanding are necessary for interpreting spectra, and progress in understanding corresponds to increasing knowledge of experimental and implicit chemical variables. Undergraduate participants exhibited uninformed bidirectional processing of all information, whereas doctoral participants exhibited informed unidirectional processing of relevant information. These findings imply the community can support novices’ development of expertise by cultivating relevant understanding and encouraging use of informed interpretation strategies, including preliminary evaluation of relevant variables, prediction of expected spectral features, and search for complementary data across spectra. These findings also provide insight into the potential relationship between conceptual understanding and information possessing across scientific representations.

Life science and medical professionals have called for undergraduate life science and pre-medical students to gain a stronger grounding in the content and methods of the physical sciences. In response, a community of physics educators have developed reformed Introductory Physics for the Life Sciences (IPLS) courses. While these courses have demonstrated success in increasing student appreciation of the value of physics for the life sciences, little work has been done to determine whether IPLS courses better prepare life science students to use and apply physical reasoning in later life science contexts. We report findings from a longitudinal interdisciplinary study in which we compare reasoning exhibited by students with and without IPLS on tasks administered in a biology senior capstone course. In particular, we observe differences in student work on a diffusion task that are correlated with prior or concurrent enrollment in IPLS. We find that IPLS students are more likely than non-IPLS students to reason quantitatively about diffusive phenomena and to successfully coordinate between multiple representations of diffusive processes, even up to two years after taking the IPLS course. These skills reflect competencies developed in the IPLS curriculum. We position these findings within the broader context of our longitudinal study of the impact of IPLS on student work in later biology and chemistry environments.

A fundamental conundrum of teaching and studying logic is that logic refers to content-general understandings while student reasoning is usually highly content-specific. When mathematics courses teach logic in de-contextualized ways, students may often fail to see how it segues with their reasoning about particular topics. In a series of teaching experiments with undergraduate students we have found a productive way to help students reinvent basic principles of logic by comparing mathematical statements across contexts. This process allows us to observe how students create content-general ways of thinking from their content-specific reasoning. The design of the activities draws heavily on the Realistic Mathematics Education tradition of guided reinvention and emergent models. I shall share in the talk some brief insights about how I have learned to revise the formulation of the logic we teach to be more compatible with and responsive to students’ ways of reasoning about mathematical language and categories. In particular, mathematical statements of the form “if then” are essentially always universally quantified (“for all”), though that language is often suppressed in mathematical text. Helping students reason specifically about sets of objects has proven challenging and yet highly productive. I shall also reflect on the importance of helping students construct productive ways of reasoning about negative categories (“not a rectangle” or “not a multiple of 6”).

The biological field is increasingly interdisciplinary and requires students to build individual concepts into complex understanding. It is important to understand how best to support students in this process and provide them with the tools necessary to succeed. One way for students to consider their own understanding, and determine what steps to take next, is by engaging in metacognition. Students engaging in metacognition have better learning outcomes, and past work has shown students can be supported to become more metacognitive. However, development of more advanced and targeted supports depends on a greater understanding of how metacognition develops as students learn increasingly complex material. In the current study, we examined students’ development of metacognition and conceptual understanding from their first introductory biology course through each of the five required core courses for the biology major. Data collection consisted of surveys, semester grades, and interviews with a subset of students. We performed statistical analysis on the metacognitive score generated from each survey and grades collected at the end of each semester. We performed qualitative trend and case study analysis on the open-ended survey questions and on the transcripts from the semi-structured interviews. We found stark differences in metacognition engagement between introductory students and seniors. We also found interesting patterns of development among those students we followed through multiple semesters. Results from this study will help to structure how future scaffolds and instructional tools are created and utilized to best support students in learning and understanding scientific concepts across disciplines.

Despite evidence that student-centered pedagogy benefits engineering students, studies show that instructors are slow to incorporate these practices into their teaching. The biomedical engineering (BME) instructional incubator (I2) is a two-semester course sequence that engages students, faculty, and postdocs in a course that demonstrates effective instruction and asks teams of student-instructors to develop a 1-credit module for early career BME students. Student-instructors then have the opportunity to teach the module with guidance from an engineering education faculty member the following semester. The I2 employs principles of situated learning to develop a community of practice that values responsive, student-centered instruction. Our study examined how I2 participants demonstrated responsive, student-centered instruction in their modules by analyzing reflective responses to a post-I2 experience survey. We coded responses for “implemented” and “proposed” changes to the modules and the corresponding “reasons for change.” After an inductive round of coding, themes emerged consistent with the curricular elements in Lattuca and Stark’s (2009) academic plan, which conceptualizes curriculum development as a process of shaping the educational environment based on institutional, departmental, and external influences. Our findings suggest that the I2 supports a community of educators developing and executing student-centered, responsive instructional practice. Implications for how this might transfer to other contexts are discussed.

In organic chemistry, learners often find it difficult to derive implicit properties of functional groups involved in mechanisms. This is necessary, for example, to make appropriate predictions about the kinetics of a reaction. Linking explicit and implicit information or deriving causal relationships poses significant challenges to students and requires specific instructions to help them solve mechanistic problems. Yet, evidence-based instructional practices to promote mechanistic reasoning in organic chemistry are limited. Scaffolded case comparisons used in interview studies have been shown to foster students’ reasoning and to facilitate the processing of case comparisons in organic chemistry. However, it is still unknown if students with a differing level of concept knowledge equally profit from such a scaffold. To address this, we conducted a study with 18 participants in an Organic Chemistry II course who were asked to solve a scaffolded case comparison individually. Prior to this task, each learner completed a concept knowledge test and a cognitive ability test in paper-pencil format, as well as a second concept knowledge test after the task. Students’ working sheets were analyzed qualitatively, to determine the reasoning patterns that emerged in students’ usage of the scaffold, and quantitatively, to relate these patterns to students’ level of concept knowledge. Every student of the cohort could be assigned to one of five reasoning patterns, which differ in aspects of multivariate and causal reasoning. We illustrate how these reasoning patterns allow to diagnose students’ understanding, as well as design adaptive tasks.

Co-author: Megan Nagel, Associate Professor of Chemistry, Penn State Greater Allegheny

Researchers in physics education have recently been applying dual-process theories of reasoning and decision-making (DPToR) as a guide to inform the development of research-based instructional materials. This approach is particularly well-suited for tasks and topics for which a strong incorrect intuitive model interferes with a student’s ability to successfully apply their conceptual understanding. In this talk, we will describe an interdiscipinary research project designed to identify topics and questions in the general chemistry curriculum for which we expect that an intervention approach rooted in DPToR would help students to improve their reasoning. This project adapts a novel tool from the physics education research literature known as the reasoning chain construction task. In this tool, students are provided with a limited number of true statements from which they must generate a reasoning chain to support their answer to a question. We will illustrate our approach and the utility of this tool with results from a task on the ideal gas law.

* This material is based upon work supported by the National Science Foundation under Grant Nos. DUE-1821390, DUE-1821123, DUE-1821400, DUE-1821511, and DUE-1821561.

Code reading is an important skill in programming. Inspired by the linearity that people exhibit while natural language text reading, we designed local and global gaze-based measures to characterize linearity (left-to-right and top-to-bottom) in reading source code. Unlike natural language text, source code is executable and requires a specific reading approach. To validate these measures, we compared the eye movements of novice and expert programmers who were asked to read and comprehend short snippets of natural language text and Java programs. Our results show that novices read source code less linearly than natural language text. Moreover, experts read code less linearly than novices. These findings indicate that there are specific differences between reading natural language and source code, and suggest that non-linear reading skills increase with expertise.

Reading skills are often a precursor to problem-solving skills. Unlike natural language text, source code is executable and requires a specific reading approach. It has been shown that one of the effective ways to improve skill acquisition is to cue visual attention of novices to the locations that experts attend while performing a task. If such intervention should be adopted in teaching program comprehension skills, knowledge about the expert behavior and significant differences between expert and novice programmers is very helpful, so that interventions can concentrate on these differences. Instructors could use tools like these to investigate the effectiveness of new pedagogies. Novice programmers could monitor their own progress and judge whether they have achieved personalized learning goals.

Directions for breakout discussion rooms
The session keynote and concurrent talks were selected to stimulate cross-disciplinary thinking. The group discussions following each session are intended to help you:

• Integrate threads across the various talks
• Share your own interests and research related to the session theme
• Consider strategies for addressing the theme within your teaching
• Make connections to people in other disciplines
• Conceptualize emerging areas for future research or programming

Group facilitators
We ask that participants engage in respectful dialogue and take care to allow everyone a chance to participate. To facilitate this process, we have identified a number of “facilitators” who will help guide the discussion. If you are in a room without a designated facilitator, please select one based on whoever has a first name starting with the earliest letter in the alphabet (and who has not previously been a facilitator). The facilitator and participants may wish to select from the following prompts as starting points for their discussions.

Potential discussion prompts
What was the most interesting perspective that could be applied to a different discipline?
What can we take from these talks that would advance theory? What were the major research insights?
How might instructors in your discipline (or a different discipline) apply this research in the classroom?
What was a new methodology that could be applied to a different discipline?
Were consistent themes apparent across talks? How did the talks compare or contrast in terms of their approaches and findings?
What other research efforts exist that might help inform this theme?
What are the most pressing areas in need of research related to the session theme?
What challenges and opportunities exist for conducting cross-disciplinary research related to the session theme?